Integrate mountpoint health check with external monitoring systems like Zabbix (configurable)

[achimnol]

depends on #1843, #1845

• Add an internal documentation (or update the installer) to configure an integrated monitoring system (e.g., Zabbix) to call the monitoring alarm API as defined in Add a manager endpoint to publish monitoring alarm events #1845.
  • e.g., POST http://{manager-addr}/monitor/alarm?reason=storage-unmounted&data={encoded-vfolder-host-dir-list}&src={agent-id}&token={some-preshared-key}
• manager: Upon receiving an "storage-unmount" MonitoringAlarmEvent (refs Add a manager endpoint to publish monitoring alarm events #1845):
  • Turn off the designated agent's schedulable attribute.
  • Add the agent status history item (refs Add status_history to agents to keep track of status changes and reasons #1843) with the reason "schedulable-off".
  • Let's provide an admin option to automatically set schedulable on when recovery is detected. We could use "storage-remount" as the reason in the MonitoringAlarmEevent.
• manager: Add an etcd configuration key config/manager/monitor/auto-exclude-agent-upon-storage-mount-failures to enable/disable the above behavior.